open-webui/backend/open_webui/retrieval/loaders
Classic298 81f5b8adf8
feat: support MinerU 4.x self-hosted servers (#32422)
MinerU 4.0 dropped the parsing interface its self-hosted server had in 3.x, so the local MinerU engine failed with a 404 on every document when pointed at a 4.x server. Open WebUI now detects a 4.x server and talks to it through its new interface, while 3.x servers keep working as before. On 4.x, `tier` in the MinerU parameters picks the parsing quality (flash, basic, standard or advanced) and `enable_ocr` forces OCR; 3.x-only settings such as `language` and `model_version` are not sent, since 4.x refuses them. 4.x also returns every image as base64 text inside the document, which would otherwise be indexed as text, so images are left out of the extracted content.

Fixes #31516
2026-10-14 23:54:41 +04:00
..
datalab_marker.py fix: Datalab Marker extraction fails on installs outside the Docker image (#32164) 2026-10-10 20:00:02 +04:00
external_document.py feat: add {{USER_GROUPS}} and {{USER_GROUP_IDS}} placeholders for custom forwarded headers (#27236) 2026-07-26 23:21:26 -04:00
external_web.py refac 2026-07-27 19:39:36 -04:00
local.py refac 2026-09-12 20:09:21 -04:00
main.py fix: Docling citations show the wrong page number when a PDF has blank pages (#32203) 2026-10-10 21:39:46 +04:00
microsoft_web_iq.py refac 2026-08-25 15:47:55 -04:00
mineru.py feat: support MinerU 4.x self-hosted servers (#32422) 2026-10-14 23:54:41 +04:00
mistral.py refactor: remove the unreachable async half of the Mistral loader and an unused Datalab helper (#28839) 2026-08-20 13:14:14 -07:00
paddleocr_vl.py perf: build info log messages lazily so raising the log level actually saves work (#27837) 2026-08-02 15:39:10 -05:00
pdf.py refac 2026-09-12 20:09:21 -04:00
tavily.py refac 2026-08-24 17:56:11 -04:00
youtube.py fix(retrieval): report why a URL could not be read instead of blaming the knowledge base (#28362) 2026-08-10 19:18:04 -06:00